ESTRO 2024 - Abstract Book

S2247

Clinical - Upper GI

ESTRO 2024

2387

Mini-Oral

Early regression index (ERI) as response predictor in esophageal cancer

Claudio A Fiorino 1 , Diego Palumbo 2 , Martina Mori 1 , Gabriele Palazzo 1 , Alessandro Pellegrini 2 , Luca Albarello 3 , Alfonso Belardo 1 , Sara Broggi 1 , Carla Canevari 4 , Andrea Cossu 5 , Anna Damascelli 2 , Ugo Elmore 5 , Elena Mazza 6,7 , Paolo Passoni 8 , Francesco Puccetti 5,7 , Najla Slim 8 , Stephanie Steidler 2 , Antonella Del Vecchio 1 , Nadia G Di Muzio 8,7 , Arturo Chiti 4,7 , Riccardo Rosati 5 , Francesco De Cobelli 2,7 1 IRCCS San Raffaele Scientific Institute, Medical Physics, Milan, Italy. 2 IRCCS San Raffaele Scientific Institute, Radiology, Milan, Italy. 3 IRCCS San Raffaele Scientific Institute, Pathology, Milan, Italy. 4 IRCCS San Raffaele Scientific Institute, Nuclear Medicine, Milan, Italy. 5 IRCCS San Raffaele Scientific Institute, Gastrointestinal Surgery, Milan, Italy. 6 IRCCS San Raffaele Scientific Institute, Psychiatry and Clinical Psychobiology, Milan, Italy. 7 Vita&Salute University, Medicine&Surgery, Milan, Italy. 8 IRCCS San Raffaele Scientific Institute, Radiotherapy, Milan, Italy

Purpose/Objective:

To prospectively assess tumor regression due to neoadjuvant chemo-radiotherapy (nCRT) for locally advanced esophageal cancer based on MRI. The ability of a previously introduced early regression index (ERI [1-3]) in early predicting complete response based on volume regression measurements was investigated.

Material/Methods:

ESCAPE is a single center prospective study conducted at our Institute and data were collected after Ethics Committee approval: all patients signed a study specific informed consent. From January 2020 to May 2023, consecutive patients with biopsy-proven potentially resectable esophageal cancer (adenocarcinoma or squamous cell carcinoma) scheduled to receive nCRT according to the CROSS scheme were prospectively enrolled. In addition to standard of care imaging, a fully integrated hybrid PET-MRI was performed at three time points. Specifically, in all cases, evaluation has been performed i) within two weeks before the beginning of nCRT (tpre) and then repeated ii) after the beginning of the third chemotherapy weekly cycle (i.e.: at radiotherapy fractions 11-14, tmid) and iii) after nCRT (one-two months after the end of radiotherapy), 8-12 weeks before surgery (tpost). nCRT delivered 41.4Gy/23fr with concurrent carboplatin and paclitaxel. Patients with complete pathological response after surgery (pCR) were considered. For patients that skipped surgery, complete clinical response (cCR) was assessed if patients showed no local relapse at the last follow-up, at least 1 year after nCRT. GTV volumes were delineated on T2w MRI images by two observers (Vpre, Vmid, Vpost) and inter-observer variability was assessed in terms of DICE index. ERI (ERImid=-ln[(1 – (Vmid/Vpre))Vpre]; ERIpost=-ln[(1 – (Vpost/Vpre))Vpost]) and other volume regression parameters at tmid and tpost were tested as predictors of complete response (pCR+cCR).

Results:

At the time of analysis, 30 patients were enrolled. Five patients were excluded (not compliance to protocol procedures (n=1), withdrawal of informed consent (n=1), tumor progression (n=2), nCRT toxicity (n=1)). Complete